People who were less physically active in their leisure time were more likely to report problems with memory, learning and concentration after retiring, according to a long-running Finnish cohort study published in Archives of Public Health. The study also found that patterns of leisure-time physical activity were broadly similar across occupational classes, with one notable exception: among participants from low occupational classes, even the most active group showed a decline in activity as they approached and entered statutory retirement.

The research drew on the Helsinki Health Study, a longitudinal cohort that at its first phase focused on employees of the City of Helsinki. Surveys were repeated across five phases between 2000 and 2022, reaching a total of 8,960 participants. For the analyses, the researchers used a final analytic sample of 4,059 participants, who had a mean age of 51.50 years at the first phase. This design allowed the team to follow activity levels over many years, spanning the transition from working life into statutory retirement.

Subjective cognitive decline was assessed with questions covering memory, learning ability and concentration, and participants were classified as either “Poor” or “Good” on this measure. Occupational class at the first phase was categorised as low, for example manual employees, or high, for example professionals. The researchers identified trajectories of activity using group-based trajectory modelling.

The modelling revealed distinct trajectories representing different levels of leisure-time physical activity and different patterns of change over the follow-up period. Most trajectories were stable and characterised by relatively high levels of activity both before and after statutory retirement, and these patterns appeared broadly similar in the low and high occupational classes. The exception was the highest activity trajectory among participants with a low occupational class, which showed a decline over the period studied.

The central finding concerned cognition. In binomial logistic regression analyses, participants in lower activity trajectories had a higher likelihood of subjective cognitive decline in both occupational classes. The association was particularly clear for learning ability: membership of the lowest trajectory group was associated with poorer learning, with average marginal effects of 22 percentage points (95% CI: 14–30) in the low occupational class and 18 percentage points (95% CI: 11–25) in the high occupational class. In practical terms, being in the least active trajectory was linked to a substantially higher probability of reporting poor learning ability after retirement in both groups.

The authors conclude that higher levels of leisure-time physical activity are associated with better cognitive health during the retirement transition in both low and high occupational class groups. Because the study is observational, it shows an association between trajectory membership and subjective cognitive decline rather than proving that physical activity protects cognition, and the cognitive outcome is based on self-reported memory, learning and concentration rather than clinical testing. The cohort consists of City of Helsinki employees, which should be kept in mind when considering how the findings generalise to other populations. The article was published open access on 20 October 2026.

Subject of Research: Trajectories of leisure-time physical activity across occupational classes and their association with subjective cognitive decline around statutory retirement

Article Title: Trajectories of leisure-time physical activity across occupational classes and their associations with subjective cognitive decline in relation to statutory retirement

Article References: Oikarinen, N., Etholén, A., Järvensivu, L., Päivärinne, V., Lallukka, T., & Rahkonen, O. (2026). Trajectories of leisure-time physical activity across occupational classes and their associations with subjective cognitive decline in relation to statutory retirement. Archives of Public Health. https://doi.org/10.1186/s13690-026-02120-w

Image Credits: AI Generated

DOI: 10.1186/s13690-026-02120-w

Keywords: leisure-time physical activity, subjective cognitive decline, retirement, occupational class, Helsinki Health Study, longitudinal cohort, MET-hours, memory, learning ability, trajectory modelling, public health, Trajectories